Machine learning approach for graphical model-based analysis of energy-aware growth control in plant factories

Yu Fujimoto, Saya Murakami, Nanae Kaneko, Hideki Fuchikami, Toshirou Hattori, Yasuhiro Hayashi

研究成果: Article

抄録

In recent decades, there has been a gradual penetration of plant factories achieving semiautomated crop cultivation. However, efficient energy utilization, as well as quality control of crops, are very important factors with regard to sustainable operation. Operating parameters, such as room temperature, affect not only the quality of crops but also the electric power required to realize the target operation while being influenced by the environment outside the plant. Therefore, a methodology is needed to analyze and interpret the relationships among these manipulated variables, exogenous variables, crop quality, and the amount of required electric power. Constructing a directed acyclic graph composed of regression models is an attractive approach for such analysis; however, the relationships can possibly be nonlinear, so the direct application of existing analytic approaches will not be appropriate. In this paper, we propose a methodology for relationship analysis among variables based on the directed acyclic graphs while identifying the linearity/nonlinearity in their relationships. In general, the construction of such a graphical model has computational issues, especially when the number of variables is large, and the risk of overfitting. The proposed method utilizes the idea of sparse regularization, which has been actively discussed in the field of machine learning, for realizing the automatic identification of linearity/nonlinearity between variables and screening redundant candidate structures; this approach relaxes the computational complexity issue and controls the risk of overfitting. As a case study, the proposed method is applied to a dataset collected from a real-world cultivation system in a plant factory to discuss its usefulness.

元の言語English
記事番号8663284
ページ(範囲)32183-32196
ページ数14
ジャーナルIEEE Access
7
DOI
出版物ステータスPublished - 2019 1 1

Fingerprint

Crops
Industrial plants
Learning systems
Control nonlinearities
Quality control
Computational complexity
Identification (control systems)
Screening
Energy utilization
Temperature

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

これを引用

Machine learning approach for graphical model-based analysis of energy-aware growth control in plant factories. / Fujimoto, Yu; Murakami, Saya; Kaneko, Nanae; Fuchikami, Hideki; Hattori, Toshirou; Hayashi, Yasuhiro.

:: IEEE Access, 巻 7, 8663284, 01.01.2019, p. 32183-32196.

研究成果: Article

Fujimoto, Yu ; Murakami, Saya ; Kaneko, Nanae ; Fuchikami, Hideki ; Hattori, Toshirou ; Hayashi, Yasuhiro. / Machine learning approach for graphical model-based analysis of energy-aware growth control in plant factories. :: IEEE Access. 2019 ; 巻 7. pp. 32183-32196.
@article{4716643ba7fd4ed5b41a9af67be7a966,
title = "Machine learning approach for graphical model-based analysis of energy-aware growth control in plant factories",
abstract = "In recent decades, there has been a gradual penetration of plant factories achieving semiautomated crop cultivation. However, efficient energy utilization, as well as quality control of crops, are very important factors with regard to sustainable operation. Operating parameters, such as room temperature, affect not only the quality of crops but also the electric power required to realize the target operation while being influenced by the environment outside the plant. Therefore, a methodology is needed to analyze and interpret the relationships among these manipulated variables, exogenous variables, crop quality, and the amount of required electric power. Constructing a directed acyclic graph composed of regression models is an attractive approach for such analysis; however, the relationships can possibly be nonlinear, so the direct application of existing analytic approaches will not be appropriate. In this paper, we propose a methodology for relationship analysis among variables based on the directed acyclic graphs while identifying the linearity/nonlinearity in their relationships. In general, the construction of such a graphical model has computational issues, especially when the number of variables is large, and the risk of overfitting. The proposed method utilizes the idea of sparse regularization, which has been actively discussed in the field of machine learning, for realizing the automatic identification of linearity/nonlinearity between variables and screening redundant candidate structures; this approach relaxes the computational complexity issue and controls the risk of overfitting. As a case study, the proposed method is applied to a dataset collected from a real-world cultivation system in a plant factory to discuss its usefulness.",
keywords = "Analysis of plant data, directed graphical model, energy-aware plant growth control, identification of linearity/nonlinearity, overlap group lasso, plant factory, sparse partially linear model",
author = "Yu Fujimoto and Saya Murakami and Nanae Kaneko and Hideki Fuchikami and Toshirou Hattori and Yasuhiro Hayashi",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/ACCESS.2019.2903830",
language = "English",
volume = "7",
pages = "32183--32196",
journal = "IEEE Access",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Machine learning approach for graphical model-based analysis of energy-aware growth control in plant factories

AU - Fujimoto, Yu

AU - Murakami, Saya

AU - Kaneko, Nanae

AU - Fuchikami, Hideki

AU - Hattori, Toshirou

AU - Hayashi, Yasuhiro

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In recent decades, there has been a gradual penetration of plant factories achieving semiautomated crop cultivation. However, efficient energy utilization, as well as quality control of crops, are very important factors with regard to sustainable operation. Operating parameters, such as room temperature, affect not only the quality of crops but also the electric power required to realize the target operation while being influenced by the environment outside the plant. Therefore, a methodology is needed to analyze and interpret the relationships among these manipulated variables, exogenous variables, crop quality, and the amount of required electric power. Constructing a directed acyclic graph composed of regression models is an attractive approach for such analysis; however, the relationships can possibly be nonlinear, so the direct application of existing analytic approaches will not be appropriate. In this paper, we propose a methodology for relationship analysis among variables based on the directed acyclic graphs while identifying the linearity/nonlinearity in their relationships. In general, the construction of such a graphical model has computational issues, especially when the number of variables is large, and the risk of overfitting. The proposed method utilizes the idea of sparse regularization, which has been actively discussed in the field of machine learning, for realizing the automatic identification of linearity/nonlinearity between variables and screening redundant candidate structures; this approach relaxes the computational complexity issue and controls the risk of overfitting. As a case study, the proposed method is applied to a dataset collected from a real-world cultivation system in a plant factory to discuss its usefulness.

AB - In recent decades, there has been a gradual penetration of plant factories achieving semiautomated crop cultivation. However, efficient energy utilization, as well as quality control of crops, are very important factors with regard to sustainable operation. Operating parameters, such as room temperature, affect not only the quality of crops but also the electric power required to realize the target operation while being influenced by the environment outside the plant. Therefore, a methodology is needed to analyze and interpret the relationships among these manipulated variables, exogenous variables, crop quality, and the amount of required electric power. Constructing a directed acyclic graph composed of regression models is an attractive approach for such analysis; however, the relationships can possibly be nonlinear, so the direct application of existing analytic approaches will not be appropriate. In this paper, we propose a methodology for relationship analysis among variables based on the directed acyclic graphs while identifying the linearity/nonlinearity in their relationships. In general, the construction of such a graphical model has computational issues, especially when the number of variables is large, and the risk of overfitting. The proposed method utilizes the idea of sparse regularization, which has been actively discussed in the field of machine learning, for realizing the automatic identification of linearity/nonlinearity between variables and screening redundant candidate structures; this approach relaxes the computational complexity issue and controls the risk of overfitting. As a case study, the proposed method is applied to a dataset collected from a real-world cultivation system in a plant factory to discuss its usefulness.

KW - Analysis of plant data

KW - directed graphical model

KW - energy-aware plant growth control

KW - identification of linearity/nonlinearity

KW - overlap group lasso

KW - plant factory

KW - sparse partially linear model

UR - http://www.scopus.com/inward/record.url?scp=85063642854&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063642854&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2019.2903830

DO - 10.1109/ACCESS.2019.2903830

M3 - Article

VL - 7

SP - 32183

EP - 32196

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

M1 - 8663284

ER -